Chapter Ocean Wind Energy Technologies in Modern Electric ...

Chapter

Ocean Wind Energy Technologiesin Modern Electric Networks:Opportunity and ChallengesFoad H. Gandoman, Abdollah Ahmadi,Shady H.E. Abdel Aleem, Masoud Ardeshiri,Ali Esmaeel Nezhad, Joeri Van Mierlo and Maitane Berecibar

Abstract

Wind energy is one of the most important sources of energy in the world. Inrecent decades, wind as one of the massive marine energy resources in the oceanto produce electricity has been used. This chapter introduces a comprehensiveoverview of the efficient ocean wind energy technologies, and the global windenergies in both offshore and onshore sides are discussed. Also, the classification ofglobal ocean wind energy resources is presented. Moreover, different componentsof a wind farm offshore as well as the technologies used in them are investigated.Possible layouts regarding the foundation of an offshore wind turbine, floatingoffshore, as well as the operation of wind farms in the shallow and deep location ofthe ocean are studied. Finally, the offshore wind power plant challenges aredescribed.

Keywords: ocean wind energy, offshore wind energy conversion, offshorerenewable energy, power transmission, offshore wind turbines, active and reactivepowers

1. Ocean wind energy resource

The increasing demand for electric power, the limited availability of fossil fuels,and increased environmental pollution have made it essential to Integra CleanEnergy sources such as wind in our energy systems. On the other hand, the shortageof drought and varieties of geographical possibilities justifies the approach todeveloping offshore wind farms.

The offshore wind farms are wind turbines that are built several kilometersoffshore in the ocean or sea for more efficient utilization of wind energy. Althoughthis method is already very costly, increasing technological advances in turbinesmaterials and bases, composite structures, as well as the construction ofmultimegawatt generators accelerate the deployment of offshore wind farms andmake them a huge part of future energy production [1–3].

This chapter starts with introduction to the wind energy resource. After that, theglobal ocean wind energy resource is presented. Ocean wind energy technologies

1

are explained in the next section. Then, the possible structure of offshore windturbine is considered. Finally, challenges of offshore wind power are discoursed.

One of the renewable energy resources that can be used to generate electricity isocean wind. Two-thirds of the Earth’s surface is water; this potential could be usedto generate electricity in different parts of the planet. However, using the oceanwind source varies depending on the geographical conditions and seasons. Cur-rently, the UK has the world’s leading outsourcing of 51% of its offshore powerplants. Denmark with 21% of the total offshore power plants in the world is in thesecond place. Other countries such as the USA, the Netherlands, Belgium, China,and Japan are also active in this field [4]. Figure 1 shows the items that need to beconsidered in the market for the production and sale of electricity through oceanwind energy. The most critical issues regarding the wind ocean project are founda-tion type and water depth which are based on geographic information in the region.

1.1 Global ocean wind energy resource

1.1.1 Ocean wind energy in Asia

Over the past few decades, large countries like China and Japan have been usingocean wind energy for electricity production. According to wind energy reports atthe World Wind Energy Council, China’s investment in this area is more than thetotal European Union and is about 3.4 GW [5].

East Asia has a high potential for exploiting ocean wind energy, and many pro-jects in this area, specifically in China, South Korea, and Japan, have been carriedout and implemented. Among Asian countries, China has more shares in the use ofwind energy. China was the first country in the Asia-Pacific region which used windocean energy. Table 1 shows several offshore wind energy projects in Asia [5, 6].

China had installed more than 3.4 GW of ocean wind capacity at the end of 2016and should end up with around 900 GW more by the end of 2030. According to

Figure 1.Essential issues in assessing ocean wind power plant.

2

Advances in Modelling and Control of Wind and Hydrogenerators

China’s Five-Year Plan, five gigawatts will be added to the country’s electricity gridby 2020 [4]. Taiwan also has the potential to produce electricity from ocean windenergy, and according to the plan of the Ministry of Energy, by 2025, it will be thetop among the active countries in Asia that use this energy [7]. Today there are onlytwo ocean turbines operating in the country, and two projects in a total of 320 MWare due to be installed by 2020. Japan has only 61 MW of installed energy for oceanwind energy by the end of 2016, given that it has more access to the ocean. SouthKorea has so far only mustered a couple of ocean wind prototypes and a singledemonstration project, in a total of 35 MW [7].

1.1.2 Ocean wind energy in Europe

The European continent has many potentials for electric power generationthrough ocean wind energy. It is revealed that this energy will play an importantrole to produce electricity for Europe in the future [8].

The Netherlands started to work on offshore wind farms after Denmark [9].Also, two ocean wind farms were built up in the Netherlands with two differentcapacity levels of 108 MW and 120 MW in 2006. Two new ocean wind farms werebuild up in Sweden by 2001 and 2002. Ireland constructed its first ocean wind farmin 2004 wind turbines of 3.5 MW. Moreover, Germany’s first ocean wind farm wasconstructed with 20,000 MW capacity. The UK used ocean wind energy by 2000with 3.8 MW capacity. France started to use ocean wind energy in 2005, but theconstruction of wind power plants for economic reasons was postponed to 2009.Table 2 shows several offshore wind energy projects in Europe [8–10].

According to the EWEA1, The European countries target is determining 20%of its power from sustainable sources by 2030. EWEA has set an objective toachieve 40 GW and 150 GW of ocean wind energy by 2020 and 2030, respectively.Additionally, through 2030, EWEA estimates yearly establishments of ocean

Country Planned capacity Project name

1 China 1.5 MW Bohai Suizhong, LiaoDong Bay

2 China 20 MW Dongshan Island

3 China 50 MW Hebei

4 China 100 MW Nan’ao, southeast of Guangdong

5 China 25 MW Shanghai Dong Mai

6 China 102 MW Shanghai Dong Mai

7 China 100 MW Fengxian No. 1

8 China 300 MW Fengxian No. 2

9 China 400 MW Nanhui

10 China 200 MW Hengsha

11 Hong Kong 200 MW Hong Kong offshore

12 Japan 2 � 600 KW Setana, Hokkaido

13 South Korea 500 MW Limjado, Jeonnam Province

14 Taiwan 4 MW Ferry

Table 1.Several ocean wind energy projects in Asia [5–7].

1 European Wind Energy Association.

3

Ocean Wind Energy Technologies in Modern Electric Networks: Opportunity and ChallengesDOI: http://dx.doi.org/10.5772/intechopen.88715

wind energy will be equal to 13,700 MW. Ocean wind energy will support 13.9% oftotal EU demand [10].

1.1.3 Ocean wind energy in Africa

Africa’s wind energy resources are focused along the coastal area and mainlandshelves. These regions ordinarily have high onshore and offshore wind energypossibilities. In 2004, the African Development Bank investigated to create a windatlas of Africa and create a quantitative guide of wind speeds over the Africancontinents [11]. Outcomes from the investigation showed that Africa’s best windenergy is found in countries adjusted along the western, northern, eastern, andsouthern shores of the African continent. The special cases are landlocked countriessuch as Chad and Ethiopia where the topographical highlights of the land areresponsible for the high wind speeds in some high-elevation zones. Additionally,according to research conducted in 2007, eight countries (Egypt, Somalia, Maurita-nia, Sudan, Libya, Chad, Kenya, and Madagascar) have high potential for onshorewind energy and five countries (Mozambique, Tanzania, Angola, South Africa, andNamibia) have high potential for ocean wind energy [11].

1.1.4 Ocean wind energy in America

The United States has vast ocean-wide areas such the Great Lakes, Hawaii,Alaska, and Gulf Coast with potential to use offshore wind energy to produceelectricity [12]. As a result, the US Department of Energy’s, Wind Energy Technol-ogies Office has conducted many studies on various technologies to facilitate elec-tricity generation from wind.

According to the US Department of Energy, the USA will have 3 GW, 22 GW,and 86 GW of ocean wind by 2020, 2030, and 2050, respectively. Therefore, theUSA will utilize 5.5% of its accessible ocean wind resources. The US Bureau ofEnergy anticipated ocean wind improvement along both the Gulf of Mexico and


1 Denmark 4.95 MW Vindeby

2 The Netherlands 2 MW Lely

3 Denmark 40 MW Middelgrunden

4 Denmark 160 MW Horns Rev

5 The UK 60 MW North Hoyle

6 The UK 60 MW Scroby Sands

7 The Netherlands 108 MW Egmond ann Zee

8 Sweden 110 MW Lillgrund

9 Netherland 120 MW Princess Amalia

10 The UK 90 MW Inner Dowsing

11 Germany 2.5 MW Breitling

12 Ireland 25.2 MW Arklow Bank

13 Sweden 10 MW Yttre Stengrund

14 Italy 0.08 MW Brindisi

Table 2.Several offshore wind energy projects in Europe [8–10].

4


West and East Coasts, in the Great Lakes, by 2050. Table 3 shows several oceanwind energy projects in the USA [12, 13].

1.2 Ocean wind energy technologies

In the last few decades, the technology used to exploit ocean wind energy forgenerating electricity has been increased day by day. These technologies depend onthe geographic region, the depth of water, and the wind speed. The main parts ofthe ocean wind power plant include [14]:

1.Tower

2.Blades

3.Gearbox

4.Power electronic components

5.Transmission system (cables)

6.Generator

1.2.1 Ocean wind turbine blade technology

The blades of the ocean wind turbine are one of the unique parts of the windturbine structure. They have unique mechanical and aerodynamic characteristics.Moreover, the technology of manufacturing wind turbine blades has undergonenew developments in both process fields and materials used in them. As a result,manufacturers of these blades are trying to optimize mechanical properties in theaerodynamic blades by design optimization and using new materials. Compositefibers and various resins, including various materials, have been used in the pro-duction of wind turbine blade rotor.

Extraction of kinetic energy from the wind is carried out by wind turbine blades.Therefore, having an optimal design to get the most energy out of the wind is veryimportant. Wind turbine blade design consists of two main parts. In the first step,the aerodynamic design is performed to achieve the required power ratingaccording to the turbine wind turbine and to obtain the highest electric power


1 The USA 30 MW Block Island (RI)

2 The USA 468 MW Cape Wind (MA)

3 The USA 500 MW US Wind (MD)

4 The USA 1000 MW DONG Energy (MA)

5 The USA 1000 MW Deepwater ONE (RI/MA)

6 The USA 2000 MW Dominion Virginia Power (VA)

7 The USA 450 MW Blue Water’s Mid-Atlantic Wind Park

8 The USA 400 MW Offshore MW (MA)

Table 3.Several ocean wind energy projects in the USA [12, 13].

5


factor. In the second step, changes should be made to the blades so that the amountof aerodynamic noise generated by the blades is within the permissible range.

1.2.2 Ocean wind turbine tower technology

Wind turbine tower is the most significant, heaviest, and most expensive part ofthe wind turbine. Regarding the safety level, its failure can cause the entire windturbine to fail. Proper design of the wind turbine tower significantly reduces thecost and increases the life of the wind turbine.

The tower is a cone-shaped steel structure with four segments mounted oneach other by screws and flanges. The tower design includes the following mainsteps [15]:

• Connector analysis

• Shell strength analysis (static analysis, bending, and aging)

• Vibration analysis

• Design and selection of all internal components of the tower (entrance door,ladder, elevator, and internal platforms)

1.2.3 Ocean wind turbine gearbox technology

The purpose of using a gearbox is to transmit relatively large forces, change thetorque or change the direction of rotation, or change the angle of the rotation axis.Gearboxes are increasing the nominal speed of a rotor from a small amount (a fewtens of rpm) to a high value (at a rate of several hundred or several thousand rpm),which is suitable for triggering a standard generator. Ideally, the resultant value isconstant in the torque at the inlet and outlet of the gearbox, but due to the energylosses in a mechanical device, torque is reduced in the output axis. In a windturbine, the power transfer from the main rotor to the generator is usually done inthree ways [16].

1.2.3.1 Direct drive transfer gearbox

In this method, the transmission is not used from the gearbox, and the torque isdirectly inputted from the main rotor to the generator. So, instead of using thegearbox and extending the main rotor, a generator with more poles is used. Toaccommodate more poles on the generator, the diameter should be increased. Oneof the benefits of using this design is to reduce the cost of the gearbox maintenanceas well as reduce gearbox shocks and increase efficiency.

1.2.3.2 Power transfer by conventional gearbox (parallel shaft)

In this method, the power output is transmitted by a conventional gearbox to thegenerator. The gears used in this gearbox can be simple or spiral. To increase theupper period, it may be possible to use two or more rounds. In parallel shaftgearbox, the bearings are used to keep the gear shaft on the main body. In this typeof gearbox, a helical gear is used, so in addition to radial force, the bearings mustalso bear a large axial force.

6


1.2.3.3 Transmission by planetary gearbox

Using this type of gearbox is very common in wind turbines. The gearbox usesthree types of gears, the Sun gear in the middle, the Planetary gear, and the Ringgear, which is an internal gear. The division of force into planetary planes, reducedgearbox size, reduced slip between the gear and the planet, and increased efficiencyrelative to other gearboxes are benefits of the planetary gearbox.

1.2.4 Ocean wind turbine energy conversion systems

Wind turbine blades convert wind energy into rotational energy in the trans-mission system, and in the next step, the generator transfers the turbine’s energy tothe grid. The most types of electric generator part in wind turbines are asynchro-nous and synchronous generators. Also, DC generators have been used for somesmaller turbines. Table 4 shows the different structure of ocean wind energyconversion [17].

In general, generators used to convert energy from offshore wind farms can bedivided into two main categories, which can be described as follows:

Asynchronous generators Synchronous generators

Table 4.Different topologies of wind energy conversion systems.

7


• Synchronous generator

• Induction generator

A. Squirrel cage induction generator

B.Wound rotor induction generator

1.2.5 Ocean wind turbine power transmission technology

The construction of wind farms requires a large space. Therefore, the best optionfor removing this limitation is the construction of these power plants in the ocean.Because the distance between offshore wind farms and the distribution network ishigh, it is better to use high-voltage direct current (HVDC) to transmit energyproduced. The suitable transmission for ocean wind farms based on HVDC isline-commutated HVDC and voltage source converter (VSC-HVDC) [18–20]. If thelength of the transmission lines is less than 50 kilometers, the use of high voltagealternating current transmission systems is not recommended. In the HVDC

Figure 2.(a) Structure of thyristor-HVDC and (b) structure of IGBT-HVDC.

Figure 3.Different types of AC transmission line of ocean wind farm.

8


technology to control active power, reactive power, and voltage thyristors isreplacing with IGBT (Figure 2a and b).

The HVAC transmission network is divided into different types, each with itsadvantages and disadvantages. The low-frequency AC transmission (LFACT) andfractional frequency transmission system (FFTS) are new transmission systemswhich have been used for the wind farm as a solution to cover the disadvantages ofconventional AC transmission line. Figure 3 shows a different type of AC trans-mission line of the ocean wind farm.

2. Active and reactive power control in the ocean wind energy system

Recently, one of the most eco-friendly and accessible renewable energy sources(RESs) being utilized all across the world is wind energy [21]. Considering thebeneficial characteristic of clean energy resources such as flexible control and regu-lation, RESs’ expansion programs and sustainable growth to reduce the greenhousegas emissions are the main purposes of European Commission in Energy Road map2050 [22]. Regarding the necessity of employing renewable energies, and also theremarkable global growth in the use of such energy sources, wind energy hasimportant advantages such as zero-emission energy production and low operatingcosts. In spite of these benefits, there is a severe uncertainty in predicting windspeed as a big challenge for such system’s integration [23]. Various machines areused in wind turbines including permanent magnet synchronous generator(PMSG), squirrel cage induction generator (SCIG), and doubly fed induction gen-erator (DFIG). Hence, the DFIG is one of the most significant types of generatorsbeing installed in wind turbines. In DFIG, both stator and rotor are connected to themain power grid directly and by power electronic converters, respectively [24–28].As it can be seen in Figure 1, the typical circuit of a doubly fed induction generatorDFIG is specified, by taking into consideration the several important parts includingmaximum power point tracking (MPPT), rotor and grid side controllers and powerelectronic converters, and pulse wide modulation PWM. Both windings of the statorand rotor of the induction machine are connected to the grid directly and by powerconverters, respectively. To active and reactive power control, several differenttypes of controller have been evaluated in pervious, research works. For instance,Ref. [29] uses sliding mode (SM) and PI controller, for controlling the stability andalso to track reference power and remove fluctuations or active and reactive pow-ers’ disturbances. After pointing out the performance of the PI controller in theoutput, the obtained results are compared with a SM controller. It has been revealedthat the PI controller has a more desired performance than SM controller from theoutput responses of the controllers’ point of view. In [30], a neural-type-1 fuzzycontroller is used to produced powers of wind turbine; the derived results arecompared to a PI controller, specifying that the neural-fuzzy controller operatesbetter than the PI controller. Refs. [31, 32] have used fuzzy-PI and sliding mode andalso robust fuzzy-sliding mode controller (F-SMC) to control and better manage thegenerated Ps and Qs of the wind turbine system output. Hence, comparing ofcontrollers’ performance indicates that the output responses of the fuzzy-proportional integration controller are better than sliding mode controller. All thecontrollers used in the literature have been used to enhance the stability and toeliminate the fluctuation and disturbances as well as improve the reference powertracking. Furthermore, each control method has modified the pervious approachesand also improved the active and reactive powers extraction of DFIG. Another typeof controller employed in DFIG generator is the fuzzy-sliding mode controller. Infact, the F-SMC is a controller which finds the best numerical values for the scaling

9


factor of the sliding mode. Due to the presence of fuzzy controller, all gain coeffi-cient value is more accurate. Wind turbine operation is affected by uncertainty dueto the uncertainty in the wind speed forecasts; thus the abovementioned controllersare not appropriate as they are not capable of dealing with severe uncertainties. It isnoteworthy that the PI controller is point to point and type-1 fuzzy controller is ableto only cover a small range of uncertainty in its outputs. By considering the highuncertainty in the wind speed, it is needed to design a controller with the capabilityof covering severe uncertainties. According to the control methods used in theliterature, and also in order to improve and enhance the power control (Ps & Qs) inwind turbines systems, a new control method based on type-2 fuzzy logic controllaws is presented. In this research, all parameters and equations are linear enablingthe system designer to use MAMDANI inference system. The T2FL controller is asuitable alternative for the controlling of the powers in DFIG to deal with severeuncertainty.

2.1 Mathematical model of wind energy conversion system

Indeed, the DFIG is an induction machine which its stator and rotor is interfacedto the main power grid. The connection of the stator and electrical energy trans-mission from its windings to the main power grid is carried out by 3-phase powertransmission lines from wind turbine to the grid. On the other hand, the rotorwounds of the induction machine are fed through the AC-DC/DC-AC back to backpower converters, which received the electrical energy from 3-phase power trans-mission lines between stator and the main power grid. In this section, a generalmodel is presented within a dynamic framework, by considering the variationparameters of the wind turbine based on induction machine. In this regard, themathematical and dynamic equations are expressed as currents, voltages, and fluxrelations in q-d-0 reference frame and also the electromagnetic and mechanicaltorque. Figure 1 illustrates the wind turbine circuit loop general operation processwith controllers, in particular the rotor-side one [30, 33].

2.1.1 Universal model of the wind turbine

By taking into consideration the rate of wind speed at the different times andalso due to the coefficient λwind:βð Þ, the mechanical power transferred from windturbine blades to the rotor shaft can be written as follows [27, 34]:

Pt ¼ π ρ R2V3wind Cp λwind:βð Þ

2(1)

wherein ρ is air density and R, Vwind, andCp λwind:βð Þ are the radius of theturbine rotor, wind speed, and the coefficient of the power capture, respectively.Hence, the power capture coefficient Cp λwind:βð Þ can be defined as follows [33, 34]:

Cp λwind:βð Þ ¼ C1C2

λiC3β� C4

� �e�C5λi þ�C6λ (2)

where λwind and β are denoted as ratio of the tip speed and blades pitch angle,respectively. Therefore, λwind can be calculated according to the following:

λwind ¼ ωtotal:RU

¼ ωr:G:RU

(3)

10


With regard to the relation (1), the input mechanical torque of the wind turbineis obtained by following:

Tmech ¼ Ptotal

ωtotal¼

π ρ R2V3wind Cp λwind:βð Þ

2

ωtotal(4)

2.1.2 DFIG mathematical model theory

Generally, there are three states or significant vector models, which are used todesign an induction machine called as Park, Clark, and Concordia. Hence, 2-phasevector model reference frame of PARK is considered for designing the inductiongenerator. On the other hand, all relationships of doubly fed induction machine areused, is under the dynamic equations in 3-phase (a, b, c) transformation framesystem to (q-d-0) 2-phase reference frame. The main parameters of formulas canbe described as stator and rotor voltage, current, flux in q-d-0 system, and alsoelectromagnetic and mechanical torque [35–38].

Figures 4 and 5 represent the vector diagram of the DFIG PARK’s model and2-phase reference frame, respectively. In these figures the conversion system from(a, b, c) reference rotating frame to (q-d-0) reference frame has been shown [32].

DFIG’s voltage equations of the stator and rotor in 2-phase reference system aredefined as follows [34, 37]:

Ud Stator ¼ RStatorId Stator þ 1ωbase

dψd Stator

dt� ωωbase

ψq Stator

Uq Stator ¼ RStatorIq Stator þ 1ωbase

dψq Stator

dtþ ωωbase

ψd Stator

8>>><>>>: (5)

Ud Rotor ¼ RRotorId Rotor þ 1ωbase

dψd Rotor

dt� ω� ωRotor

ωbaseψq Rotor

Ud Rotor ¼ RRotorIq Rotor þ 1ωbase

dψq Rotor

dtþ ω� ωRotor

ωbaseψd Rotor

8>>><>>>: (6)

where U:R:I:ψ is characterized as the stator and rotor voltages, resistances, andcurrents, fluxes in d-q reference frame, and also the rotor base speed, rotor speed,and angular speed is denoted by ωb:ωr:ω respectively.

Figure 4.Doubly fed induction generator PARK’s model [32].

11


Also, the induction machine’s currents are in 2-phase reference frame, which arecalculated as follows:

Id Stator ¼ ψd Stator

Xl Stator� ψmd

Xl Stator

Iq Stator ¼ψq Stator

Xl Stator� ψmq

Xl Stator

8>><>>: (7)

Id Rotor ¼ ψd Rotor

Xl Rotor� ψmd

Xl Rotor

Iq Rotor ¼ψq Rotor

Xl Rotor� ψmq

Xl Rotor

8>><>>: (8)

By considering current Eqs. (7) and (8), the parametersψ :ψMagnetic:Xl Stator:Xl Rotor are known as the, flux, magnetizing flux, and stator androtor leakage reactance in q-d two-dimensional vector space. Hence, under suchcondition the fluxes of the stator and rotor windings can be formulated as follows:

ψq Stator ¼ ωbase

ð RStator ψmq � ψq Stator

� �Xl stator

þ Vq Stator

0@

1Adt

ψd Stator ¼ ωbase

ðRStator ψmd � ψd Statorð Þ

Xl Statorþ Vd Stator

� �dt

8>>>>><>>>>>:

(9)

ψq Rotor ¼ ωbase

ð RRotorðψmq � ψq RotorÞXl rotor

þ ωRotor

ωbaseψd Rotor þ Vq Rotor

� �dt

ψd Rotor ¼ ωbase

ð RRotor ðψmd � ψd RotorÞXl rotor

þ ωRotor

ωbaseψq Rotor � Vd Rotor

� �dt

8>>><>>>: (10)

In addition, by taking into consideration the current and flux parameters in2-phase reference frames of the rotor and stator of the induction generator, thetorques can be obtained as (11) and (12):

Telectromagnetic ¼1:5 P ψd Stator Iq Stator � ψq StatorId Stator

� �2ωbase

(11)

Figure 5.The DFIG in the 2-phase reference frame [32].

12


where P, ωbase are the number of pole and base speed of the induction generator’srotor, respectively, which are considered in the calculation of electromagnetictorque. Due to the relation (11), the mechanical torque can be written as follows:

TMechanical ¼ Jddt

ωrm � Tem þ Tdamp (12)

In (12), the parameters Tdamp, J are defined as the damping torque and rotormoment of inertia, considering the rotor speed variations.

2.1.3 Ps & Qs power control process in DFIG-based WECS

The active and reactive power control strategy is that the fuzzy controller byreceiving error signals and derivative of the error and also, after a control process,the signals generated through the controller is delivered to the power electronicconverters and DFIG’s rotor and to the active and reactive output power control ofthe wind turbine. The relationships of Ps & Qs powers are defined as follows[40, 41]:

Ps ¼ 1Xl Magnetic

ψd Statorð Þ þ XMagnetic

Xl Stator1:5 Id Rotor Vd Rotorð Þ

� �(13)

Qs ¼1

Xl Magneticψq Stator

� �þ XMagnetic

Xl Stator1:5 Iq Rotor Vq Stator� � �

(14)

2.2 Controller design based on the type-2 fuzzy logic theory T-2FLC

2.2.1 Type-2 fuzzy controller statement

T-2 fuzzy controller is a developed controller in that its operation strategy isunder the uncertainty. Therefore, this type of controller is appropriate for systemswith high uncertainty such as wind or solar power plants, which the generation isexactly under the pure uncertainty. The performance of the T2FLC because ofcovering a large scale of high uncertainty to control the wind turbine parameters ismore desired than the T1 fuzzy or another controller technique. Structurally, bothT1 and T2 fuzzy controllers are the same, but with this difference in the interiorstructure of T2FLC, due to the presence of uncertainty, there is a section as typereducer (TR). The calculation and the conversion of the type of fuzzy from type 2to type 1 are the important functions of the TR section. Depending upon thelinearity and nonlinearity of the model, two types of inference system exist inT2FLC. The MAMDANI inference system is used for the systems with linearequations, and TSK inference system is employed for the systems with nonlinearequations.

2.2.2 Extended fuzzy sets

Generally, T2 fuzzy sets are extended of type-1 fuzzy sets. On the other hand,T2FS is a fuzzy set with membership degree of fuzzy. Type-2 fuzzy sets can com-pensate the limitation of type-1 fuzzy sets in covering uncertainties as a newmethod with its specific advantages. Forasmuch as fuzzy sets are defined based onlinguistic variables, thus the T2FS is appropriate to model uncertainty processusing linguistic variable. The primary membership grade in T1FSs is a crisp numberin [0, 1], whereas the primary membership grade in T2FSs is a T1FSs in range of

13


[0, 1] and the secondary membership is a crisp number in [0, 1] as well [17, 22].Figure 6 represents T2FSs with its upper, lower, and footprint ranges of theuncertainty.

As shown in Figure 7, A T2FS consists of foot printing of uncertainty (FOU),upper membership function (UMF), lower membership function (LMF), andembedded fuzzy system wherein the FOU and embedded FS have been shown asblue lines. If all membership grades of FT2 sets are the same in the secondary part ofFT2 sets, the sets of FT2 are the internal type, otherwise the general type. Withrespect to the uncertainty in the membership function of FT2 sets, the generalconcept of type-2 fuzzy set is defined by the relations below [37, 38].

A type-2 fuzzy set is characterized in a function (H) and is described as follows:

H ¼ y:v�

:μH y:v� ��

:∀y ∈Y:∀v ∈Zy ⊆ 0:1½ � (15)

wherein

y∈Y:v∈Zy ⊆ 0:1ð Þ:0≤ μH y:v�

≤ 1 :

Figure 6.The general circuit of a DFIG and its connection to the grid.

Figure 7.Type-2 fuzzy set with FOU and embedded FS and lower and upper memberships.

14


In order to reduce the computational complexity, the interval fuzzy sets (IFSs)are proposed as an alternative to the general fuzzy sets (GFSs). Hence, a set ðeHÞ ofinterval type2 is defined as follows:

eH ¼ðy∈Y

ðv∈Zy

1y:v� Zy ⊆ 0:1½ �

eH : Y ! E:F½ � : 0≤ E≤ F≤ 1f g:

8><>: (16)

The performance of fuzzy inference engine is under the considered rules. Withseven membership functions, 49 of the commands written to form IF-THEN havebeen considered as represented in Table 1 [37–42]. The relation of footprint ofuncertainty is as follows:

FootPrintin eH� �¼ ∪∀y∈YZy ¼ y:v

�: ∪∈Zy ⊆ 0:1½ ��

(17)

According to the impacts of the uncertainty in FT2 sets, the bound of FT2 setsincludes two fuzzy type-1 set membership functions as upper membership function(UMF) and lower membership function LMF. The embedded fuzzy sets in the set ofH can be defined as follows as a eHe set [18]:

eHe ¼ðy∈Y

1v

�y

:v∈Zy (18)

2.2.3 Principle process of the type-2 fuzzy logic inference system

The general performance of T2FLC is based on rules and relationships whichhave been considered for it. As it can be seen in Figure 8, the principle process oftype-2 and type-1 fuzzy control systems are the same, but with this difference thatthe FT2 control system has a unit called fuzzy type reducer. FT2 system comprisesfive important parts in that the first part is the fuzzifier unit, while the inputs of[0, 1] interval are converted to fuzzy sets. The second part is the inference engineunit wherein; all fuzzy sets are inferred by rule base unit, simultaneously.Depending upon linearity or nonlinearity of the fuzzy control inputs and the equa-tions, the inference system in (T2FLS) control system can be MAMDANI or TSK.The next part after the inference engine taken into consideration the most impor-tant part of FT2 logic system is the rule base unit. All fuzzy inference calculationsare according to the human knowledge and written in the frame of IF-THEN.

Figure 8.Overall process of FT2 logic system.

15


The fourth section of FT2 system is the type reducer. Since the FT2 sets are based onthe uncertainty and due to the high computational burden of the fuzzy system, thisis not possible for the system output to be converted to [0, 1] directly. First, all thesets of the FT2 are converted to FT1 sets using the type reducer and then applied tothe defuzzifier unit and converted to [0, 1] in the output. The last part of the FT2system is the defuzzifier unit in that its performance is in the opposite way of thefuzzifier system and converts all fuzzy sets to [0, 1] [38–40].

By taking into consideration the significant parts of type-2 fuzzy topology, theprocess of inference is expressed in the form of mathematical, which is denoted asfollows:

EM ¼ FLower:FUpper �

(19)

subject to

FMLower ¼ min μNK

1

� �Lower

y1�

:… μNKP

� �Lower

yP� h i

(20)

FMUpper ¼ min μNK

1

� �Upper

y1�

:… μNKP

� �Upper

yP� � �

(21)

And the minimum and maximum computational of type reducer can beexpressed in the fractional functions as follows:

Zl ¼ minδi ∈ FKLower xið Þ�FK

Upper xið Þ½ �PN

i¼1 xi δiPNi¼1 δi

(22)

Zr ¼ maxδi ∈ FKLower xið Þ�FK

Upper xið Þ½ �PN

i¼1 xi δiPNi¼1 δi

(23)

Finally, defuzzification is the next step after the type reduction unit which inorder to achieve the controller’s output is done by:

Zc ¼ Zl

2þ Zr

2(24)

2.2.4 Type-2 fuzzy controller design

Generally, Figure 9 shows the process of the data inference, analyze and con-version them from crisp system input [0, 1] to the type-2 fuzzy system and againtransform to the crisp system output [0, 1]. According to the main system’s inputequations, the design of FT2 controller has been done using the FT2 toolbox. TheFT2 controller detail, such as error, change of error that is gain input (KP, KD),fuzzy inference system unit, output gain (KU) with its intervals, the number ofconsidered membership functions for inputs and output, some of its laws, and, also,the type of inference, is expressed in the form of a toolbox. Given the linearity ofthe equations of DFIG, the MAMDANI inference system with Gaussian member-ship functions has been considered for the FT2 controller [39]. The main part oftype-2 controller is the fuzzy inference (FIS) section, in which all operating levels offuzzy sets can be done by this part. Since this work is focused on the rotor-sidecontroller RSC and also by considering the presence of uncertainty in the windspeed, elimination of the oscillation (overshot), as well as stability enhancement ofthe output powers Ps & Qs of the wind turbine based on DFIG, is the principaltarget of this essay. With regard to Figure 1 that indicated the general structure of

16


the DFIG, hence, as it can be seen, the type-2 fuzzy controller input parameters arethe active and reactive powers, and its output is the voltage of the rotor in the q-d-0reference frame. The performance of fuzzy inference engine is under the consid-ered rules. With seven membership functions, 49 of the commands written to formIF-THEN have been considered as represented in Table 1 [37, 38, 40]. To betterindicate the concept of the type-2 toolbox and its application to T2FL controllerdesign, the general scheme of an interval type-2 toolbox with MAMDANI inferencestrategy and also the type of inputs and output membership functions with itsranges are depicted in Figures 5 and 7, respectively. As shown in Figure 8, thestructure of type-2 fuzzy logic controller is composed of input gains (KP, KD),type-2 fuzzy inference unit, output gain (KU), and plant as well. Indeed, the plantsection is a mathematical transfer function. Since the type-2 fuzzy inference sectionis the main part of T2 controller and on the other hand its function is directly basedon the IT2 toolbox, thus, under such conditions, it is required that all parameters’information about the type-2 controller system, such as input and output scalingfactors, number of rules and command, membership functions, and its ranges, aredefined in the toolbox as well. All sections of the T2 fuzzy logic controller are shownin Figure 10.

Notation 1: Each letter in Table 5 has a special meaning. For instance, negativebig (NB), negative medium (NM), negative small (NS), zero (ZO), positive small(PS), positive medium (PM), and positive big (PB).

As shown in Figure 11, the structure of type-2 fuzzy logic controller is composedof input gains (KP, KD), type-2 fuzzy inference unit, output gain (KU), and plant as

Figure 9.Type-2 fuzzy logic toolbox.

Figure 10.Input and output Gaussian membership functions with its intervals.

17


well. Indeed, the plant section is a mathematical transfer function. Since the type-2fuzzy inference section is the main part of T2 controller and on the other hand itsfunction is directly based on the IT2 toolbox, thus, under such conditions, it isrequired that all parameters’ information about the type-2 controller system, suchas input and output scaling factors, number of rules and command, membershipfunctions, and its ranges, are defined in the toolbox as well. All sections of the T2fuzzy logic controller are shown in Figure 11.

Each of letters in Table 1 has a special meaning. For example, negative big is themeaning of NB, and ZO is the abbreviation of zero, while the following describesthe fuzzy rules:

If error is negative big and change of error is negative big, then KU isnegative big.

2.2.5 Tuning of FT2 controller’s gains using the PSO algorithm

PSO is one of the most popular optimization algorithms which is operatedaccording to the social treatment of birds and aquatics movement. The process ofoptimization in the algorithm ends whenever using the pre-defined stop criteria[43, 44]. In this article, (PSO) algorithm is used to tuning the input and outputscaling factors of the controller. To optimize the output powers (Ps & Qs) of thewind turbine through the T2 fuzzy controller, it is required to properly tune theinput and output gains of the controller [45–47]. Under such conditions, each of theinput and output scaling factors of the type-2 controller will have a suitable number,in which its numerical amounts are determined by PSO algorithm. In the presenceof uncertainty and due to the complexity and the large number of the FT2 equa-tions, it would be very difficult or even impossible to choose an optimal number orenter values manually into the input and output gains. Accordingly, PSO algorithmhas been used in this paper to accelerate adjusting the coefficients to get the propernumber and more accurate response to regulate the input and output scaling factorsof the controller. The PSO algorithm is based on the particles’ behavior includingthe velocity and the location of particles [48–53]. Taking into consideration thegeneral structure of the PSO algorithm, the process of coefficients regulation of the

Figure 11.The main structure of the type-2 fuzzy logic controller simulation.

E/COE NB NM NS ZO PS PM PB

NB NB NB NB NM ZERO ZERO ZERO

NM NB NB NB NM ZERO ZERO ZERO

NS NB NB NB NM ZERO ZERO ZERO

ZO NM NM NM ZERO PM PM PM

PS ZERO ZERO ZERO PB PB PB PB

PM NB NM NS PM PB PB PB

PB PS PM PB PB PM PB PB

Table 5.Type-2 fuzzy rule chart.

18


FT2 controller’s input and output gains using PSO algorithm is defined in threesteps. At the first step, a general cost function is created including the names of thecontroller’s gains characteristic; the name of the main system that the type-2 fuzzycontroller is considered for, i.e., a DFIG-based wind turbine; and the sum of errorand the change of the error. In the second step, the main values such as the numberof parameters, the minimum and maximum values of the input and output of theFT2 controller gains (KP, KD, KU), the name of the cost function, the number ofmaximum iteration, as well as all parameters relating to the PSO algorithm aredefined. In the third step, the best numerical value of the FT2 controller gains isdetermined by running the PSO algorithm. By considering a larger number ofiteration loops in the algorithm to adjust the gains of the controller, the outputresponse will be improved. The general structure process of the type-2 controller’sgain regulation has been depicted in Figure 12. As shown in Figure 9 and also withthe presence of the T2 fuzzy controller in this system, the PSO algorithm adjusts allscaling factors of the T2FL controller by receiving the error and change of error(E, COE) as the input and then chooses the best value for each gain of the controller(KP, KD, KU) in the output. To better understand the optimization procedure bythe PSO algorithm, all the algorithms’steps are described as a flowchart in Figure 13.

Notation 2: Indeed, the PSO algorithm is based on the cost function for which itis intended. In order to membership functions tuning of the type-2 fuzzy controllergains, the cost function is defined as follows:

Function H= Cost Function-FCN (KP, KD, KU)Sim (‘DFIG’)H=Sum ((e. ^ 2) + (De. ^ 2))End

2.3 Simulation results

This part expresses the simulation results obtained using the presented frame-work. In this regard, the obtained results using the proposed FT2 controller arecompared with those obtained by the FT1 controller. Figure 14(a) and (b) showsthe error and change of error surfaces of the FT1 and FT2 controllers, respectively.In this regard, the FT2 controller has a smoother surface than FT1 due to thecovering uncertainty in a large and different ranges and high computational burden.This paper is focused on the power (P & Q) control using the RSC. According to thecircuit loop of doubly fed induction machine, the general power control process in

Figure 12.Tuning of fuzzy type-2 controller gain process using PSO algorithm.

19


the DFIG-based wind turbine can be stated in multiple stages. At the first stage,after entering the value of the measured power (the generated power by the DFIGin the initial moment without controller’s function), reference power is comparedby the type-2 fuzzy controller. Regarding the transfer function considered for theT2 FLC, the output signal of the controller is the rotor voltage in d-q referenceframe. Since the input of pulse wide modulation (PWM) unit is the voltage in the a,b, c reference frame, at the second stage, first the controller’s output voltage in d-qframe is converted to the a, b, c frame by a d-q to a, b, c unit transformation; afterthat the controller’s output signal will be sent to the PWM block, and at the thirdstage, the output signal of the PWM will be transferred to the rotor-side powerconverter. In the presence of uncertainty of the wind speed, the main goal of thissimulation is to show the stability of the powers on the considered referencenumerical amount, using a T2 fuzzy controller. In order to power stability on thevalue of 400 W, the reference power should be adjusted to 400. The active andreactive power output responses have been exactly stabilized at the reference

Figure 13.Flowchart of the PSO algorithm.

Figure 14.(a) and (b),The FT1 and FT2 control surface.

20


amount which is considered for the outputs of the T1 and T2 fuzzy controllers. Allthe results are depicted in Figures 15 and 16. As it can be seen in the figures, boththe powers Ps & Qs are in the stable mode after multiple overshot in the transientstate of the FT1 controller output, but in FT2 controller, the active and reactivepowers have been stabilized without any overshoot or oscillation in the transientstate. Indeed, before the active power and reactive power are stabilized in theoutput, all overshoots or disturbances are removed in the transient states by the FT2controller. In the FT2 controller due to the high computational burden, the activeand reactive powers become stabilized with more latency compared to the FT1controller. In this part, to better indicate the wind turbine and also for more

Figure 15.(a) and (b). The output results of the active power (a) and reactive power (b), controlled using the FT1controller.

Figure 16.(a) and (b). The active power (a) and reactive power (b) output control using the FT2 controller.

21


information, all the numerical values of the main parameters of the DFIG designwith its acronyms are organized in Table 6. The DFIG numerical amount chartconsists of voltage, resistance, inductance leakage of the rotor and stator, magneticinductance, moment inertia, and the number of poles.

Notation 3: Since, in this paper, the main control aim is sustainable of theoutput powers without any fluctuation or is less overshoots at the reference value;hence, the numerical value 400 W is considered for both controllers, as the refer-ence amount.

2.3.1 P & Q powers controlled using type-1 fuzzy controller

With regard to Figure 15(a) and (b), both powers Ps and Qs of the type-1controller’s output, until reaching to the stable conditions, have faced multipleoscillation (overshoots) in its transient state, and also, by taking into considerationthe presence of uncertainty, the powers are stabilized at the reference amount 400W. The FT1 controller just covers a small interval in its output, and as the perfor-mance of the main system which is based on the uncertainty, the FT1 controllerwould not be able to properly control the output powers because of the presence ofuncertainty, in which the transient states of FT1 controller output are with multiplefluctuations. However, multiple fluctuations occur in the FT1 controller’s outputmainly due to the uncertainty.

2.3.2 Ps & Q powers controlled using type-2 fuzzy controller

Figure 16 demonstrates the active and reactive powers outputs. Therefore, asshown, the performance of T2 controller is better than T1 controller; in otherwords, the powers have improved in its transient mood by considering the presenceof the uncertainty, and on the one hand, T2FLC has a smoother surface in its controlof the output powers. Since the computational burden of the mathematical theoryof the type-2 fuzzy strategy is high, the output response of the FT2 controller untilthe stable state is associated with a time delay of several seconds. Due to thecapability of the FT2 controller in covering a large range of the uncertainty, fluctu-ations have been removed, and it presents a smoother behavior in its transient state.In this paper, as previously described, with a little time delay, both P & Q poweroutputs of the FT1 and FT2 controller are stabilized at the value of 400 W. Thestability of the active and reactive powers at the reference value has been depicted

Parameters Acronyms Numerical values

Frequency (F) 60

Stator line voltage (VL rmsð Þ) 200

Stator resistance (RS) 3.35

Rotor resistance (Rr) 1.99

Stator leakage inductance (LLS) 6:94e�3

Rotor leakage inductance (LLrp) 6:94e�3

Magnetic inductance (LM) 163.73e�3

Moment inertia (J) 0.1

Number of pole (P) 4

Table 6.Numerical values of the main parameters of the DFIG with its acronyms.

22


in Figure 16(a) and (b), respectively. Ps and Q powers output response of T1FLCindicates the controlled output power stability at the pre-defined reference value inFigure 13. As specified in Figure 16(a), it has been shown that the active power isstabilized at the value of 400 W. By taking into consideration the type-1 and type-2controller functions, the major difference between them is to cover the uncertainty.The T1 controller cannot cover uncertainty, or in other words, it just controls thepowers over a small specified range, while the T2 fuzzy controller can cover theuncertainty in large scales.

Both active power and reactive power output response of the type-1 FLC indi-cate the controlled output power stability at the pre-defined reference value inFigure 16. In Figure 16(a), it has been shown that the active power is stabilized atthe value of 400 W. Regarding the functions of type-1 and type-2 fuzzy controllers,the major difference between them is to cover the uncertainty. The T1 controllercannot cover uncertainty, or in other words, it just controls the powers over a smallspecified range, while the T2 fuzzy controller can cover the uncertainty in largescales.

2.3.3 Voltages [UasUbsUcs] and currents [IasIbsIcs] of the stator

In this section, the results of 3-phase voltage and currents between DFIG’s statorand the main power grid through the power transmission lines have been investi-gated. Hence, the results of the stator 3-phase voltages are characterized in theframe of [UaUb Uc]. Also, in order to show the numerical range of sinusoidal outputof the wind turbine, the numerical value 200 V is intended for the output system.Figure 17(a) indicates that the output voltage of the DFIG’s stator is in the range of

Figure 17.(a) and (b). 3-phase voltages from grid to the stator in (a, b, c) reference frame with the sinusoidal waveform.

23


Figure 18.(a) and (b). 3-phase currents of the stator in (a, b, c) reference frame with sinusoidal waveform.

Figure 19.(a) and (b). 3-phase rotor voltage with sinusoidal waveform.

24


(�200, +200) exactly. So, the sinusoidal waveform of the three-phase voltages hasspecified in Figure 14(b) clearly. Figure 18(a) is illustrates the stator currentresponse by taking into consideration the range of (�10, +10). Generally, thewaveforms of the stator 3-phase currents are sinusoidal as well, which have beendepicted in Figure 18(b).

2.3.4 Voltages [UarUbr Ucr] of the rotor

The input of the PWM is the voltage in [arbrcr] frame. Regarding the inputvoltages of the rotor, which will be in a, b, c frame, therefore, before the delivery ofthe signal from controller to the PWM, the output of the FT2 controller that is thevoltage rotor in d-q-0 frame should be converted to a, b, c reference system. Theinterval of the rotor’s 3-phase voltage depends upon the considered value. In thispaper, the desired numerical amount is (200 V). Figure 19(a) represents the three-phase rotor output voltage that is stable in the range of [�200, +200]. The value ofthe PWM unit input voltage must be in per unit. In principle, the rotor voltagewaveform is sinusoidal which has been depicted in Figure 19(b).

3. The possible structure of an offshore wind turbine

In general, the structure of ocean wind farms can be divided into two maincategories:

1.From the perspective of the foundation used for construction

2.From the depth of water view

This chapter addresses the above issues.

3.1 Offshore wind turbine from foundation point of view

Offshore turbines are placed in the water and have more complexity to install ona turbine mounted on the land. Figure 20 shows the different foundations for oceanwind turbines. Additionally, offshore wind turbine foundation must withstandharsh condition as well. This explains the wide variety of foundation developed overthe years for offshore turbines, some more proven than others [54].

Figure 20.The different foundations for ocean wind turbines.

25


3.2 Offshore wind turbine from the depth of water point of view

The layout of offshore wind farm changes based on the geographic area, thestructure of the wind turbine, and the depth of water. The structure of waterturbines in shallow water, deepwater, and floating has been investigated.

3.2.1 Shallow water offshore turbine

For areas with a water depth of fewer than 40 m, the use of offshore wind farmsis appropriate.

Figure 21 shows the typical structure of shallow water wind turbines:

A.Gravity base

B. Mono-plie

C.Mono-caisson

D.Multi-pile

E. Multi-caisson

3.2.2 Deepwater offshore wind turbine

For areas with a water depth of more than 40 m, the use of low wind turbine isappropriate. Figure 22 shows the typical structure of deepwater wind turbines:

A.Tripod tube steel

B. Guyed tube

C. Spaceframe

D.Talisman energy concept

Figure 21.Models of ocean wind turbines in shallow water.

26


3.2.3 Floating offshore wind turbine

Floating wind turbines are constructed on a floating structure on water, which iskept in different ways on the ocean floor. This method is used in areas where it isnot possible to make a foundation for them. Figure 23 shows three types of floatingstructures in an offshore wind turbine.

1.Tension leg mooring systems

2.Catenary mooring systems

3.Ballasted catenary configuration

3.3 Offshore wind farm design

The design of offshore wind farms should be considered from three crucialpoints. Figure 24 shows the design process for a typical ocean wind turbine.

Figure 22.Models of ocean wind turbines in deepwater.

Figure 23.Three types of engineered design for anchoring floating structures.

27


4. Challenges of offshore wind power plants

In general, two types of offshore wind power plant structures are challenging,(i) fixed offshore wind turbine and (ii) floating offshore wind turbine. Also,important issues that are being considered as the current challenge in offshore windfarms are the turbine layout and the way electricity is transmitted from the ocean tothe shore [55].

Offshore installations currently consist of only a small percentage of the renew-able energy market. However, due to the advancement of technology in the designand evaluation of these types of energy resources, it is expected that much progresswill be made shortly. Offshore wind farms are in the early stages of their commer-cialization. They demand a higher cost of capital than onshore wind farms, but thiscan be compensated by higher capacity factors [56]. Offshore wind farms allowmore widespread utilization of wind energy potentials. The reason for the highercapacity factors and the possibility of more use of offshore wind energy are asfollows:

Figure 24.The design process for a typical ocean wind turbine.

28


• There are no obstacles and restrictions for installing wind turbines for offshoreturbines.

• It is possible to build sizeable enormous wind farms offshore.

• In places where the wind speed is average, it can be constructed.

• Offshore turbines have more extended and higher blades, leading to moreswept areas and higher electricity outputs.

The offshore turbines are designed for use in offshore. Due to the lack of focuson such issues as shaking and impulse, noise, and visual contamination, there is arelatively different technical path. Although the issue of increasing turbine size foroffshore turbines is a problem, this will increase the profit but also increase theoperating costs. In this regard, the changing of the design and the ability to considerconsiderations are likely to provide better conditions in the design of offshoreturbines.

Now, new turbines have a power of at least 5 MW. Therefore, a 1000 MWpower plant can be achieved by installing 200 turbines [56]. Increase in the cost ofthe offshore wind turbine installation in the sea and the transfer of energy to thecoast are most significant problems which need to be considered. Researchers arestill trying to find the ways to reduce the cost of ocean wind farm.

Another challenge in the construction of offshore wind farms is the shortage oflarge ships that can carry large and heavy parts such as turbines. Also, anotherchallenge in the field of offshore wind farms is the incentive to participate in theelectricity market. Power transfer from the plant to the power grid using suitableinfrastructures is also challenging for the use of this future energy. There are manyother challenges that need to be addressed with the availability of sufficient tech-nology in the world and companies who are active in wind turbine production [57].Currently, the number of companies specialized in this field is insufficient, and it isexpected that the number of these companies will increase shortly. Finally, toexpand the use of this energy source, the training of a specialist who can build and

Figure 25.The lifecycle of an offshore power plant.

29


operate offshore wind farms is another issue that should be addressed by electricitycompanies.

Regarding the abovementioned explanation, the major challenges of offshorewind technology are the high cost of offshore wind provision, lack of currentinfrastructure to support the fabrication such as installation, operation intercon-nection, maintenance of the system, and the challenges related to the lack of sitedata and lack of experience.

Figure 25 shows the lifecycle of an offshore power plant. According to thisdiagram, to solve the first challenge, it should be possible to reduce the impact ofthis problem in the long-term reports, with the development of industries and thereduction of installation costs and increased reliability of the system.

Nowadays, installation of ocean wind turbines requires specialized vessels,grid interconnections, purpose-built portside infrastructure, and robust underseaelectricity transmission lines will be useful from the financial point of view.Regarding the last challenge, ocean wind projects confront new and untestedallowing processes, which contributes to the uncertainty and risk faced by potentialproject developers and financiers.

Author details

Foad H. Gandoman1,2, Abdollah Ahmadi3, Shady H.E. Abdel Aleem4*,Masoud Ardeshiri5, Ali Esmaeel Nezhad6, Joeri Van Mierlo1,2 andMaitane Berecibar1,2

1 Research Group MOBI—Mobility, Logistics, and Automotive TechnologyResearch Center, Vrije Universiteit Brussel, Brussels, Belgium

2 Flanders Make, Heverlee, Belgium

3 School of Electrical Engineering and Telecommunications, The University of NewSouth Wales, Sydney, NSW, Australia

4 Mathematical, Physical and Engineering Sciences, 15th of May Higher Institute ofEngineering, Cairo, Egypt

5 Department of Electrical Engineering, Kazerun Branch, Islamic Azad University,Kazerun, Iran

6 Department of Electrical, Electronic, and Information Engineering, University ofBologna, Italy

*Address all correspondence to: [email protected]

©2020TheAuthor(s). Licensee IntechOpen.Distributed under the terms of theCreativeCommonsAttribution -NonCommercial 4.0 License (https://creativecommons.org/licenses/by-nc/4.0/),which permits use, distribution and reproduction fornon-commercial purposes, provided the original is properly cited. –NC

30


References

[1] Blaabjerg F, Ma K. Wind energysystems. Proceedings of the IEEE. 2017;105(11):2116-2131

[2] Lua X, McElroya MB, Kiviluomac J.Global potential for wind-generatedelectricity. PNAS. 2009;106(27):10933-10938

[3] Lua X, McElroya MB. Chapter 4—Global potential for wind-generatedelectricity. In:Wind Energy EngineeringA Handbook for Onshore and OffshoreWind Turbines. Elsevier; 2017. pp. 51-73

[4] Zheng CW, Li CY, Pan J, Liu MY,Xia LL. An overview of global oceanwind energy resource evaluations.Renewable and Sustainable EnergyReviews. 2016;53:1240-1251

[5] Li W, Yao H, Wang H, Wang Z.Research progresses in assessment ofChina's offshore wind energy resources.Journal of Renewable and SustainableEnergy. 2014;6(05):31-38

[6] GWEC-Global-Wind-2015-Report_April-2016_22_04, www.gwec.net

[7] Leary D, Esteban M. Recentdevelopments in offshore renewableenergy in the Asia-Pacific region. OceanDevelopment & International Law.2011;42:94-119

[8] Makridis C. Offshore wind powerresource availability and prospects: Aglobal approach. Environmental Science& Policy. 2014;6(05):31-38

[9] Mytilinou V, Kolios AJ, DiLorenzo G. A comparative multi-disciplinary policy review in windenergy developments in Europe.International Journal of SustainableEnergy. 2015;36(8):754-774

[10] The European offshore windindustry. Key Trends andStatistics 2016. windeurope.org

[11] Belward A, Bisselink B, Bódis K,Brink A, Dallemand J-F, de Roo A, et al.Renewable energies in Africa.Luxembourg: Publications Office of theEuropean Union; 2011. Available from:http://www.jrc.ec.europa.eu/

[12] Musial W, Heimiller D, Beiter P,Scott G, Draxl C. 2016 Offshore WindEnergy Resource Assessment for theUnited States. NERL, Denver WestParkway: National Renewable EnergyLaboratory; 2016

[13] Daniel J, Liu S, Pan J. Potentialimpacts and economic value of USoffshore wind. In: IEEE Power & EnergySociety General Meeting; Denver, CO.2015. pp. 1-5

[14] Sun X, Huang D, Wu G. The currentstate of offshore wind energytechnology development. Energy. 2012;41(1):298-312

[15] Li W, Zhou H, Liu H, Lin Y, Xu Q.Review on the blade design technologiesof tidal current turbine. Renewable andSustainable Energy Reviews. 2016;63:414-422

[16] Nejad AR, Guo Y, Gao Z, Moan1 T.Development of a 5 MW referencegearbox for offshore wind turbines.Wind Energy. 2016;19:1089-1106

[17] Toftevaag T. Specialization CourseELK 12. NTNU; 2012

[18] Blaabjerg F, Ma K. Future on powerelectronics for wind turbine systems.IEEE Journal of Emerging and SelectedTopics in Power Electronics. 2013;1(3):139-152

[19] Gandoman FH, Sharaf AM, AbdelAleem SHE, Jurado F. DistributedFACTS stabilization scheme for efficientutilization of distributed wind energysystems. International Transactions

31


on Electrical Energy Systems. 2017;27(11):1-20

[20] Gandoman FH, Ahmadi A,Sharaf AM, Siano P, Pou J, Hredzak B,et al. Review of FACTS technologies andapplications for power quality in smartgrids with renewable energy systems.Renewable and Sustainable EnergyReviews. 2018;82:502-514

[21] El-mohr I. Renewable energysystems. Wind Energy. 2010. Availablefrom: http://slideplayer.com/slide/10266678/

[22] Bhutto DK, Ansari JA, Chachar F,Katyara S, Soomro J. Selection ofoptimal controller for active andreactive power control of doubly fedinduction generator (DFIG). In:Computing, Mathematics andEngineering Technologies (iCoMET),2018 International Conference on.IEEE; 2018. pp. 1-5

[23] Wu C, Nian H, Pang B, Cheng P.Adaptive repetitive control of DFIG-DCsystem considering stator frequencyvariation. IEEE Transactions on PowerElectronics. 2018;34(3):3302-3312

[24] Sáiz-Marín E, Lobato E, Egido I.New challenges to wind energy voltagecontrol. Survey of recent practice andliterature review. IET Renewable PowerGeneration. 2017;12(3):267-278

[25] Parida A, Chatterjee D. IntegratedDFIG–SCIG-based wind energyconversion system equipped withimproved power generation capability.IET Generation, Transmission &Distribution. 2017;11(15):3791-3800

[26] Lin F-J, Tan K-H, Tsai C-H.Improved differential evolution-basedElman neural network controller forsquirrel-cage induction generatorsystem. IET Renewable PowerGeneration. 2016;10(7):988-1001

[27] Wang Y, Meng J, Zhang X, Xu L.Control of PMSG-based wind turbines

for system inertial response and poweroscillation damping. IEEE Transactionson Sustainable Energy. 2015;6(2):565-574

[28] Hu J, Nian H, Hu B, He Y, Zhu Z.Direct active and reactive powerregulation of DFIG using sliding-modecontrol approach. IEEE Transactions onEnergy Conversion. 2010;25(4):1028-1039

[29] Naidu NS, Singh B. Grid-interfacedDFIG-based variable speed wind energyconversion system with powersmoothening. IEEE Transactions onSustainable Energy. 2017;8(1):51-58

[30] Kouadria S, Messlem Y,Berkouk EM. Sliding mode control ofthe active and reactive power of DFIGfor variable-speed wind energyconversion system. In: Renewable andSustainable Energy Conference(IRSEC), 2015 3rd International.Marrakech, Morocco: IEEE; 2015.pp. 1-8

[31] Ranga KP, Sukumar GD,Pakkiraiah B, Rao MS. Neuro fuzzybased P s &Q S controller for DoublyFed Induction Generator with windturbine. In: Electrical Power and EnergySystems (ICEPES), InternationalConference on. Bhopal, India: IEEE;2016. pp. 139-144

[32] Mohanty A, Patra S, Ray PK. Robustfuzzy-sliding mode based UPFCcontroller for transient stability analysisin autonomous wind-diesel-PV hybridsystem. IET Generation, Transmission& Distribution. 2016;10(5):1248-1257

[33] Hamane B, Doumbia M,Bouhamida M, Benghanem M. Controlof wind turbine based on DFIG usingFuzzy-PI and Sliding Mode controllers.In: Ecological Vehicles and RenewableEnergies (EVER), 2014 NinthInternational Conference on. Monte-Carlo, Monaco: IEEE; 2014. pp. 1-8

32


[34] Li P, Wang J, Wu F, Li H. Nonlinearcontroller based on state feedbacklinearization for series-compensatedDFIG-based wind power plants tomitigate sub-synchronous controlinteraction. International Transactionson Electrical Energy Systems. 2018;29:e2682

[35] Taher SA, Dehghani Arani Z,Rahimi M, Shahidehpour M. A newapproach using combination of slidingmode control and feedback linearizationfor enhancing fault ride throughcapability of DFIG-based WT.International Transactions on ElectricalEnergy Systems. 2018;28:e2613

[36] Ashouri-Zadeh A, Toulabi M,Bahrami S, Ranjbar AM. Modification ofDFIG's active power control loop forspeed control enhancement and inertialfrequency response. IEEE Transactionson Sustainable Energy. 2017;8(4):1772-1782

[37] Ghosh S, Kamalasadan S. Anintegrated dynamic modeling andadaptive controller approach forflywheel augmented DFIG based windsystem. IEEE Transactions on PowerSystems. 2017;32(3):2161-2171

[38] Mendel JM, Wu D. Critique of “anew look at Type-2 fuzzy sets and Type-2 fuzzy logic systems”. IEEETransactions on Fuzzy Systems. 2017;25(3):725-727

[39] Raju SK, Pillai G. Design andimplementation of type-2 fuzzy logiccontroller for DFIG-based wind energysystems in distribution networks. IEEETransactions on Sustainable Energy.2016;7(1):345-353

[40] Yassin HM, Hanafy HH,Hallouda MM. Enhancement low-voltage ride through capability ofpermanent magnet synchronousgenerator-based wind turbines usinginterval type-2 fuzzy control. IETRenewable Power Generation. 2016;10(3):339-348

[41] Khooban MH, Niknam T, Sha-Sadeghi M. Speed control of electricalvehicles: A time-varying proportional–integral controller-based type-2 fuzzylogic. IET Science, Measurement &Technology. 2016;10(3):185-192

[42] Sarabakha A, Fu C, Kayacan E,Kumbasar T. Type-2 fuzzy logiccontrollers made even simpler: Fromdesign to deployment for UAVs. IEEETransactions on Industrial Electronics.2018;65(6):5069-5077

[43] Chaoui H, Khayamy M,Aljarboua AA. Adaptive interval type-2fuzzy logic control for PMSM driveswith a modified reference frame. IEEETransactions on Industrial Electronics.2017;64(5):3786-3797

[44] Rahman NHA, Zobaa AF.Integrated mutation strategy withmodified binary PSO algorithm foroptimal PMUs placement. IEEETransactions on Industrial Informatics.2017;13(6):3124-3133

[45] Wei L-X, Li X, Fan R, Sun H,Hu Z-Y. A hybrid multiobjectiveparticle swarm optimization algorithmbased on R2 indicator. IEEE Access.2018;6:14710-14721

[46] Mahdi FP, Vasant P, Abdullah-Al-Wadud M, Watada J, Kallimani V.A quantum-inspired particle swarmoptimization approach for environ-mental/economic power dispatchproblem using cubic criterion function.International Transactions on ElectricalEnergy Systems. 2018;28(3):e2497

[47] Serrano FE, Flores MA. C++ libraryfor fuzzy type-2 controller design withparticle swarm optimization tuning. In:Central American and PanamaConvention (CONCAPAN XXXV), 2015IEEE Thirty Fifth. Tegucigalpa,Honduras: IEEE; 2015. pp. 1-7

[48] Wang D, Ma N, Wei M, Liu Y.Parameters tuning of power system

33


stabilizer PSS4B using hybrid particleswarm optimization algorithm.International Transactions on ElectricalEnergy Systems. 2018;28:e2598

[49] Hurel J, Mandow A, García-CerezoA. Tuning a fuzzy controller by particleswarm optimization for an activesuspension system. In: IECON2012-38th Annual Conference on IEEEIndustrial Electronics Society. Montreal,QC, Canada: IEEE; 2012. pp. 2524-2529

[50] Lee JH, Song J-Y, Kim D-W, KimJ-W, Kim Y-J, Jung S-Y. Particle swarmoptimization algorithm with intelligentparticle number control for optimaldesign of electric machines. IEEETransactions on Industrial Electronics.2018;65(2):1791-1798

[51] Koad RB, Zobaa AF, El-Shahat A. Anovel MPPT algorithm based on particleswarm optimization for photovoltaicsystems. IEEE Transactions onSustainable Energy. 2017;8(2):468-476

[52] Tulay G, İskender İ, Erdem H.Optimal tuning of a boost PFC converterPI controller using heuristicoptimization methods. InternationalTransactions on Electrical EnergySystems. 2017;27(12):e2458

[53] Javadi MS, Esmaeel Nezhad A.Intelligent particle swarm optimizationaugmented with chaotic searchingtechnique to integrate distant energyresources. International Transactions onElectrical Energy Systems. 2017;27(12):e2447

[54] Ahmed SS. Finite element modelingof large diameter monopiles in densesand for offshore wind turbinefoundation. In: Proceedings of theASME 2015 34th InternationalConference on Ocean, Offshore andArctic Engineering, May 31–June 5,2015, Newfoundland, Canada

[55] Malhotra S. “Design andConstruction Considerations for

Offshore Wind Turbine Foundations”,Research and Innovations. New York:P D Network; 2004

[56] Oh KY, Nam W, Ryu MS, Kim J,Epureanu BI. A review of foundations ofoffshore wind energy convertors:Current status and future perspectives.Renewable and Sustainable EnergyReviews. 2018;88:16-36

[57] Santos AC, Perez JP, Diez DB,Rodríguez C. Offshore wind energy: Areview of the current status, challengesand future development in Spain.Renewable and Sustainable EnergyReviews. 2016;64:1-18

34


Chapter Ocean Wind Energy Technologies in Modern Electric ...

Documents

Transcript of Chapter Ocean Wind Energy Technologies in Modern Electric ...